help with build

[trem]

I have been having a strange problem with this amp that I cannot figure out. The schem and pictures are below. I am only using one input to one half of the first gain stage (so just one channel with just one input, no high and low). I currently do not have the Tremolo Circuit tube installed.
Briefly, the amp powers up, and seems to have "normal B+ voltages, although they are not listed anywhere. But, the amp does not pass any signal for about one minute. When the amp starts to pass signal through the speaker, there is a big change in the B+ of the phase inverter and the first preamp tube.
So, when I first power up, and no signal is passing, the cap that feeds the PI and Pre is: 275 VDC.
When signal starts to come out the speaker, that voltage drops to 247 VDC. This same thing happens at the plate of the PI and first pre tube. Those numbers are as follows.
PI plates
Pin #2- 269V and then 146V
Pin #5- 199V and then 213
Cat resistor to ground- 1.5V and then 2.5V

Preamp tube:
Pin #2- 119V and then 107V
Cat resistor- 1.2V and stays the same at 1.2V

Do you guys have any idea what could be wrong, or what I could have wired wrong that would cause this.? I have replaced a lot of the involved caps and resistors (including the filter cap) and re-flowed all of the solder joints. I cannot do anything to fix this. Any suggestions.?
Thank You

http://www.webphix.com/schematic%20h...ps/m12_6v6.pdf

http://i111.photobucket.com/albums/n...rAmpage001.jpg
http://i111.photobucket.com/albums/n...rAmpage002.jpg
http://i111.photobucket.com/albums/n...rAmpage003.jpg

[Prattacaster]

Does the amp work ? The voltages the amp settles at dont seem that wacky although i would think the PI pin2 and pin 5 would be a little closer. Does the voltage at the power tube cathodes change also?

[trem]

I am pretty sure the power tube cats and the B+ for the plates and screens are not effected by this. I will double check all of it.
Yes, the amp works after it starts to pass signal after a minute or 2.
Thanks

[redelephant]

Are the heaters using long time to get to normal op-temp? Like one minute...?
The schem shows a 22k resistor in series with the b+ for the PI and pre. A drop from 275V to 247V indicates an increase of about 1.5mA, not enuff to seem odd. What is the b+, or voltage on the output tranny?
The fact that your PI has differing plate voltages, and those voltages change the way they do, makes me wonder if there is some DC voltage creeping into one of the PI's grids turning one more on than the other?

edit, rethink: the difference on the PI's plates after (one minute) is 14volts, I am not sure that's too bad, there is always some mismatch, and with the feedback going into an offset cathode connection like this, I'd expect some difference.

When sound does go thru, does it sound ok? It may be that the heater secondary simply is a little on the low side and needs some time to heat the heaters. Since when they are cold, they have about 1/3 the resistance of when they are hot, and can pull the secondary down which in turn delays the heaters getting hot enough...???